On this page we provide a brief introduction to complement-activation related pseudoallery and its clinical relevance. For a more detailed discussion please read the article Complement activation-related pseudoallergy: A new class of drug-induced acute immune toxicity(11) published in Toxicology (download HERE in .pdf format).

A major goal in modern pharmacotechnology is to increase the therapeutic index of drugs by using nanoparticulate vehicle systems in order to ensure slow release or targeted delivery. With all great benefits, however, these innovative therapies can carry a risk for acute immune toxicity manifested in hypersensitivity reactions (HSRs), also called anaphylactic, anaphylactoid or idiosyncratic reactions (10). Some of these represent classical IgE-mediated type 1 HSRs, while many others arise without the involvement of IgE. The latter reactions cannot be predicted or diagnosed by traditional allergy tests or IgE analysis, and are thus called pseudoallergy. A portion of pseudoallergic reactions were shown to be due to activation of the complement (C) system, hence, they were distinguished within the Type I category of HSRs as C activation-related pseudoallergy (CARPA) (2-11). The symptoms of CARPA include common, as well as unique features compared to Type I reactions (Table 1).

Drugs and agents causing CARPA include radiocontrast media, liposomal drugs (Doxil, Ambisome, DaunoXome and immuno-liposomes), micellar solvents containing amphiphilic lipids (Cremophor EL, the vehicle of Taxol, and polysorbate 80, the vehicle of Taxotere and Etoposide), viral gene vectors (adenovirus) and antibody-based therapeutics, e.g., IVIG and monoclonal antibodies (mAbs) (1-12).

Table 1. Symptoms of CARPA vs. Ig-E mediated hypersensitivity
Ig-E-mediated (Type I)		CARPA
Common Symptoms
Angioedema, asthma attack, bronchospasm, chest pain, chill, choking, confusion, conjunctivitis, coughing, cyanosis, death, dermatitis, diaphoresis, dispnoea, edema, erythema, feeling of imminent death, fever, flush, headache, hypertension, hypotension, hypoxemia, low back, pain, lumbar pain, metabolic acidosis, nausea, pruritus, rash, rhinitis, shock, skin eruptions, sneezing, tachypnea, tingling sensations
Unique Symptoms
Reaction arises after repeated exposure to the allergen. Reaction is stronger upon repeated exposures. Reaction does not cease without treatment Rate is low (2%)	Reaction arises at first treatment (no prior exposure to allergen). Reaction is less severe or absent upon repeated exposures. Spontaneous resolution. Rate is high (2%, up to 40%)
Causes of death in fatal reactions: cardiac arrest, shock, multi-organ failure

The above-mentioned drugs and antibodies have been known to activate C, which leads to major rises of C3a and C5a anaphylatoxins in blood, triggering mast cells and basophils for secretory responses that underlie HSRs. The formation of C membrane attack complex (C5b-9, MAC) on these cells might also contribute to their activation (3,7,11).

Fig. 1. The scheme illustrates that liposomes, micelles, antibodies and other agents that can cause HSRs can activate C leading to the formation of anaphylatoxins, perforins and opsonins. The first two processes can trigger mast cells and basophils for releasing a variety of secondary mediators with potent physiological actions. Among these, histamine, TXA2 and PAF have been shown to be causally involved in CARPA.

Clinical relevance of CARPA:
Implications for the pharmaceutical industry

Patients undergoing diagnostic procedures or treatment with a reactogenic agent need to be extensively pre-medicated with steroids and/or antihistamines, implying extra medical care and expenses with additional health risk in some individuals (2, 11). While transient, mild reactions that occur despite pre-medication are well tolerated, and are viewed as an inconvenience that is outweighed by the benefits of intervention, severe reactions cause major anxieties, disruptions, extra expenses and, most importantly, exclusion of the patient from receiving a diagnostic procedure or promising treatment. These reactions can also be fatal in a small percentage of hypersensitive individuals, mainly those with a history of severe allergy and/or heart disease.

The relevance of CARPA for the pharmaceutical industry specialized in injectables lies in the unpredictability of these reactions, i.e., the risk that despite passing all preclinical safety tests, severe reactions could later surface. Previous examples of unexpected, product killer adverse events include the case of Jesse Gelsinger, the fist volunteer to die in a human gene therapy trial presumably due to C activation arising as a consequence of antibody response against the adenoviral vector (Bostanci, Science 2002;295: 604-5).

In another recent example, the neutrophil-reactive mAB NeutroSpec, had to be withdrawn from the market because of two deaths and other serious, life-threatening cardiopulmonary events occurring within minutes of injection of this radiographic agent. Yet as recently as in March 15, 2006, an anti-CD28 mAb (TGN1412, Parexel) sent 6 of 8 phase 1 trial volunteers to intensive care in critical condition with symptoms of major angioedema and capillary leakage, causing multiple organ failure within minutes of administration.

Although these adverse events with mAbs have not (yet) been linked to C activation, the symptoms and underlying principle of treatment (antibody binding to cells in the circulation) clearly raise the possibility of C activation triggering a catastrophic overstimulation of the immune system.

Complement activation as a predictor of hypersensitivity reactions: Preliminary proof of principle

Consistent with the causal role of C activation in liposome-induced HSRs, a recent clinical study demonstrated significant correlation between the formation of C terminal complex (SC5b-9) in blood in vivo and the presence of HSRs in patients treated with liposomal doxorubicin (Doxil).  HSRs was observed in 45% of patients (n=29), and, remarkably, in 92% of these reactor patients plasma SC5b-9 values were also significantly elevated over baseline at 10 minutes after starting the infusion (Fig. 1A).  In contrast, in the nonreactor group, only 56 % of patients showed C activation (B), and the rise of SC5b-9 was significantly smaller [10]. This study, taken together with the observation that 7 of 10 normal human sera show C activation following incubation with Doxil in vitro (Fig 1, ref. 5) provides rationale to measure SC5b-9 in vitro, as a predictive marker of a risk for HSR in vivo. 

Fig. 1 (top).  Complement activation by Doxil in cancer patients in vivo.  Blood SC5b-9 levels at 10 min after strating the infusion in patients who A) developed grade 2-3 (moderate-to-severe) HSRs or B) did not display clinical symptoms.  Statistical analysis of these data (Fisher’s exact test and Cohen’s k  statistics, testing the degree of association between laboratory and clinical reactions, and the agreement (or reproducibility) of two surveys identifying patients as reactors by clinical and laboratory criteria) indicated significant (P<0.05) relationship between C activation and HSR [10]. Furthermore, the specificity and positive predictive values of the SC5b-9 assay with regards to HSR were remarkably high with SC5b-9 readings exceeding 2-4-fold the upper threshold of the normal range (dashed lines in Fig. 1A and B) (10)

Fig. 2. (bottom) C activation by Doxil in vitro in 10 different normal human sera.  The drug was incubated with the sera for 30 min at 37 degrees C and SC5b-9 was measured using Quidel’s TCC ELISA (1,5,8)

The sC5b-9 test

Freshly prepared serum from healthy volunteers or patients to be tested are incubated with the test drug for 30-45 min, followed by measurement of SC5b-9 levels using SC5b-9 ELISA [1,2,5,8,10]. Significant elevation over PBS baseline indicates C activation, which is taken as an indicator of a risk for reactogenicity in vivo.

Prediction of CARPA using a porcine model

Pigs provide a useful model for liposome-induced CARPA, as minute amounts of reactogenic liposomes cause C activation with consequent dramatic cardiovascular and laboratory abnormalities that mimic some of the human symptoms (3,4,12).  Thus, SeroScience also offers the porcine CARPA test as an alternative, large-animal in vivo model for predicting the potential of drugs to cause CARPA.  In our procedure (3,4,8), pigs are injected i.v. with milligrams of the test drug, and the ensuing changes in systemic and pulmonary blood pressure (SAP, PAP), cardiac output (CO) and pulmonary end-tidal CO2 partial pressure (pCO2) are recorded, along with the ECG changes and plasma thromboxane B2 (TXB2) levels, the latter taken as an indirect marker of anaphylatoxin release in blood [3].  Figs. 3 and 4 illustrate the hemodynamic, respiratory and ECG changes that are observed in pigs following the injection of different liposomes with different C activating potencies.  Based on the severity of hemodynamic and ECG changes, the distress reaction, and, hence, the risk of CARPA, can be quantified on an arbitrary scale called cardiopulmonary abnormality score (CAS) (12).

Figs. 3 (top) and 4 (bottom)

Hemodynamic, cardiopulmonary and ECG changes in pigs injected i.v. with Doxil. The abrupt massive hypotension is associated with pulmonary hypertension, declines of cardiac output and end tidal CO2 (Fig. 3/B), along with tachyarrhythmia (3/C), bradyarrhythmia (Fig4/A) or ventricular fibrillation (Fig. 4/B) (12).

Home | Company Profile | Background | Services | References | Contact Us